Learning to Search More Efficiently from Experience: A Multi-heuristic Approach
Abstract
Learning from experience can significantly improve the performance of search based planners, especially for challenging problems like high-dimensional planning. Experience Graph (E-Graph) is a recently developed framework that encodes experiences, obtained from solving instances in the past, into a single bounded-admissible heuristic, and uses it to guide the search. While the E-Graph approach was shown to be very useful for repetitive problems, it suffers from two issues. First, computing the E-Graph heuristic is time consuming as it maintains the bounded admissibility constraints. Second, a single heuristic can get stuck in a local minimum, and thereby, degrade the performance. In this work, we present an alternative approach to improving the runtime of search from experience, based on a recently developed search algorithm Multi-heuristic A* (MHA*). This framework provides an improvement over the E-Graph planner for two reasons: a) MHA* uses multiple heuristics simultaneously to explore the search space, which reduces the probability of getting stuck in a local minimum, and b) the heuristics in MHA* can be arbitrarily inadmissible, which makes it very easy to compute them. The paper describes the framework, explains how to compute these (inadmissible) heuristics through offline and online processing and presents experimental analysis on two domains, motion planning for a 6D planar arm and large sliding tile puzzles.
BibTeX
@conference{Aine-2015-109748,author = {Sandip Aine and Charupriya Sharma and Maxim Likhachev},
title = {Learning to Search More Efficiently from Experience: A Multi-heuristic Approach},
booktitle = {Proceedings of 8th International Symposium on Combinatorial Search (SoCS '15)},
year = {2015},
month = {June},
pages = {141 - 145},
}